Guys,

I was wondering if I could get a decent reply to my questions based on my past experiences of using Fluent.

Firstly, why on earth the solution does not converge properly when it is started on a Second Order Upwind and QUICK scheme on a hybrid mesh, however it converges perfectly when you switch to a first order upwind scheme. I used to think there used to be a problem with my mesh or something and I have just realised its not the case! I wonder why?

Secondly, say for external aerodynamics applications on wings, the Cp plot over the wing is different to each other when solved using the Spalart Allmaras and RnG K-Epsilon model and a Y+=30 to 45. Again I wonder which one is right?

I would be most grateful if anyone can help me understand what is happening.

Thanks a lot in advance.

Alan

Why "high order scheme" are less stable than first order scheme?....you will find the answer in any CFD book...

But to sum up, high order scheme are known to be more subject to numerical instability, first order scheme are usually subject to numerical diffusivity which has the tendancy to stabilize the scheme.

But a high order scheme is also able to capture more physics, as natural instability in the system.

But how to distinguish numerical and real instability ?, this is the typical work for a CFD man. Work on the grid, as much as you can (BL, and use structured grid,...etc)... verify your far field Bondary conditions...the list is long...

You can find more detailed advises:

http://www.cfd-online.com/Forum/flue...cgi?read=35319

and http://www.cfd-online.com/Forum/flue...cgi?read=34904